The focus of this proposal is proteins that interact with DNA sequences located at or near a DNA end, using adenovirus DNA replication as a model. Human adenoviruses cause respiratory infections in humans and are tumorigenic in rats and other rodents. Within the region of the genome that encodes two adenoviral DNA replication proteins (early region 2B), the predicted amino acid sequence is 80% conserved among the adenovirus serotypes known. In vitro adenovirus DNA replication requires a DNA end near to a DNA binding sequence in order for initiation to occur. Other genetic systems (such as transposition) may also require such an interaction. Our general goal is to study the events leading to the formation of an initiation complex required for DNA replication: to identify the protein domains that interact with the DNA, to map the contact points between DNA and protein, to determine the factors that influence nucleotide binding, and to study the importance of the DNA ends for initiation. The low abundance replication polypeptides will be overproduced in E. coli by recombinant DNA techniques for eventual physical characterization. DNA sequencing and computer modeling will be used to study how diverged peptide domains in the polypeptides affect secondary structure. Unusual conformations (cruciforms, left-handed DNA) will be sought within the adenovirus DNA ends. A filter binding assay will be developed to measure binding of the polypeptides to various DNA substrates during formation of the initiation complex. Long range goals include mutagenesis of the genes encoding in the replication polypeptides and crystallization of the polypeptides for structure analysis.